RESEARCH Open Access

Placental lipase expression in pregnanciescomplicated by preeclampsia: a case–controlstudyHelen L. Barrett1,2,3* , Marta H. Kubala1, Katherin Scholz Romero1, Kerina J. Denny5, Trent M. Woodruff5,H. David McIntyre3,4, Leonie K. Callaway2,3 and Marloes Dekker Nitert1,3

Abstract

Background: Preeclampsia (PE) is associated with maternal and neonatal morbidity and mortality. In PE, thephysiological hyperlipidaemia of pregnancy is exaggerated. The purpose of this study was to examine theexpression of adipose triglyceride lipase (ATGL), hormone sensitive lipase (HSL), lipoprotein lipase (LPL) andendothelial lipase (EL) in pregnancies complicated by PE.

Methods: Placentae were collected from 16 women with PE and 20 women with uncomplicated pregnanciesmatched for maternal prepregnancy BMI and gestational age of delivery. Gene and protein expression of theplacental lipases were measured by Q-PCR and Western blot. DNA methylation of the promoter of LPL wasassessed by bisulfite sequencing. Lipase localisation and activity were analysed.

Results: Gene expression of all lipases was significantly reduced, as was HSL protein level in women with PE. Alllipases were localised to trophoblasts and endothelial cells in PE and control placentae. There was no difference inmethylation of the LPL promoter between PE and control placentae. Lipase activity was not altered in placentaefrom women with PE.

Conclusion: These results suggest that the decreased placental lipase gene but not protein expression or lipaseactivity, which is associated with late-onset PE is not a major contributor to the abnormal lipids seen in PE.

Keywords: Preeclampsia, Intrauterine growth restriction, Lipase, Placenta, Pregnancy

BackgroundPreeclampsia (PE) occurs in ~ 5 % of pregnancies in thedeveloped world. During pregnancy, PE is associatedwith maternal multiorgan dysfunction, placental abrup-tion and poor fetal growth. In the longer term, PE pre-dicts maternal hypertension and carries an increasedmaternal cardiovascular and renal morbidity [1, 2]. Oneadverse infant outcome associated with PE is intrauter-ine growth restriction (IUGR), a failure of the infant toreach its full potential growth [3]. IUGR is associatedwith perinatal morbidity and mortality and also with

hypertension and cardiovascular disease later in life forthe infant [4, 5].In PE, most studies report exaggerated and early ma-

ternal gestational hyperlipidaemia [6–8], with markedhypertriglyceridemia, higher very low density lipoprotein(VLDL) concentrations [9] and higher levels of small,dense low density lipoprotein (LDL) [10]. There is also arise in maternal free fatty acid (FFA) concentrations abovenormal pregnancy levels [11]. This excessive increase inmaternal lipids is thought to contribute to endothelial dys-function, one of the hallmarks of PE [12, 13]. PE is also as-sociated with abnormalities in lipid oxidation, which maybe part of the underlying pathophysiology of the condition[12, 13]. Women with a history of PE demonstrate persist-ent abnormalities in lipids postpartum [14, 15].The placenta supplies fatty acids (FFA) and cholesterol

to the infant. While FFAs can diffuse across the placenta,

* Correspondence: h.barrett@uq.edu.au1UQ Centre for Clinical Research, The University of Queensland, Herston,QLD, Australia2Obstetric Medicine, Royal Brisbane and Women’s Hospital, Herston, QLD,AustraliaFull list of author information is available at the end of the article

© 2015 Barrett et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 DOI 10.1186/s12958-015-0098-9

most lipids require active placental transport throughthe activity of lipoprotein receptors, lipases and fattyacid binding proteins [16, 17]. Abnormal maternal lipidsor altered placental lipid processing could contribute tothe altered infant growth and change in cord lipopro-teins seen in PE. The expression and activity of lipopro-tein lipase (LPL) has been variably found to be higher,lower or unchanged in placentae from women with PEor IUGR compared to uncomplicated pregnancy[18–26]. Endothelial lipase (EL) gene expression hasbeen reported to be decreased in IUGR [19] but has notbeen analysed in PE. The expression and localization ofthe intracellular lipases adipose triglyceride lipase(ATGL) and hormone sensitive lipase (HSL) have notpreviously been examined in placentae from womenwith pregnancies complicated by PE or IUGR.The current study aims to examine placental lipase ex-

pression in late-onset PE. Placental gene, protein expres-sion and localization of ATGL, HSL, EL and LPL wasanalysed in placentae from women with PE and IUGRand uncomplicated pregnancy. Furthermore, DNA methy-lation of the promoter of LPL was investigated and overalllipase activity was examined.

MethodsSubjectsPregnant women in the third trimester were recruitedfrom a tertiary general and obstetric hospital. All womengave written informed consent. Permission for the studywas granted by the Human Research Ethics Committeesof the Royal Brisbane and Women’s Hospital and TheUniversity of Queensland. Diagnosis of PE was definedby current Society of Obstetric Medicine, Australia NewZealand guidelines research definition [27]. Participantswere matched for maternal BMI, which was calculatedfrom a recorded early pregnancy weight in kg divided bythe squared height in meters. The customized birth cen-tile was calculated with the online calculator gestation.net(www.gestation.net). Small for gestational age (SGA)infants were defined as adjusted birth weight centile <10th. Placental tissue was collected immediately post-

delivery, sampled randomly (~ 1 cm3) but away fromareas of infarction or calcification, snap-frozen in liquidnitrogen and kept at -800 C until analysis. In addition,1 cm3 samples of placenta for paraffin embedding werewashed in PBS, placed into 4 % paraformaldehyde for48 hours and kept in a saturated sucrose solution untilembedding.

Ethics approvalThis study was approved by the Human Research EthicsCommittees of the Royal Brisbane and Women’s Hos-pital (HREC/08/ARBW/16: 19/01/2009) and The Uni-versity of Queensland (2009000115: 04/02/2009).

RNA isolation and quantitative real-time PCRPlacental tissue was lyzed by violent shaking for 2 × 2minutes at 30 Hz with a 5 mm stainless steel bead in aTissueLyser (Qiagen, Chadstone, VIC, Australia). mRNAwas isolated from placenta with the Allprep RNA/DNAextraction mini kit (Qiagen). RNA was quantified byNanodrop and all samples had 260/280 ratios > 1.8.750 ng mRNA was reverse transcribed to cDNA withthe QuantiTect reverse transcription kit (Qiagen) usingan equal mixture of oligodT and random primers. Quan-titative real-time PCR was performed on 18.75 ng ofcDNA with 300 nM of primers and iTaq universal SYBRgreen mastermix (Bio-Rad, Gladesville, NSW, Australia)on an iQ5 PCR machine (BioRad). The PCR protocolconsisted of 1 cycle at 95 °C for 10 min, 40 cycles of 95 °C for 15 sec and 59 °C for 1 min followed by dissoci-ation curve analysis. Primers unique for the target geneand covering exon-exon junctions were designed withprimerBLAST. The primer sequences are presented inTable 1. To adjust for potential differences in cellularcomposition of the placental samples, gene expressionwas normalized to the geometric mean of expression ofthe housekeeping gene TATA-box binding protein(TBP), cytokeratin 7 (CK7) as a marker for trophoblastcells, CD34 (CD34) for endothelial cells and desmin(DES) for smooth muscle cells. The analysis was also

Table 1 Primer sequences

Gene name Forward primer Reverse primer

LPL 5′-TGGATCGCTCCACTTTGACC 5′-GGGCTTCGGACTGGTAAACA

LPIG 5′-GTCCAGCCCCTGCTATCTCA 5′-CCTTTTCAAACTGACCCTTGCC

LIPE 5′-CACATTAGACCCAGAAGATGCC 5′-GGCAGCGAAACTTGACAGTG

PNPLA2 5′-TGCCCACTTTGTGTGTATGTG 5′-CCAGGAGTGCGACGCT

CK7 5′- CCGTGCGCTCTGCCTATGGGG 5′- GCTCCAGAAACCGCACCTTGTCGAT

CD34 5′- CCACAGGAGAAAGGCTGGGCGA 5′- AGCCCCTCGGTTCACACTGGC

DES 5′- TCCGAGAAACCAGCCCTGAGCAA 5′- GTGGCCTCACTGACGACCTCCC

TBP 5′-GGGCACCACTCCACTGTAC 5′-CTGTTCTTCACTCTTGGCTCCT

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 2 of 9

performed normalizing to the expression of TBP onlyyielding similar results.

Protein expressionPlacenta were lysed with a RIPA buffer consisting of50 mM Tris, 1 % Triton-X, 0.1 % SDS, 0.5 % DOC,150 mM NaCl, and protease inhibitor cocktail (Roche,Applied Science, VIC, Australia). Tissue was disruptedby violent shaking for 2 × 2 minutes at 30 Hz with a5 mm stainless steel bead in a TissueLyser (Qiagen).After lysis, the sample was centrifruged for 10 min at 4 °C and the protein content in the supernatant determinedby bicinchoninic acid assay (Sigma-Aldrich, Castle Hill,NSW, Australia). 30 μg of protein was loaded onto a 4–12 % gradient NuPAGE® Bis-Tris gel (Life Technologies,Mulgrave, VIC, Australia), transferred onto a polyvinyli-dene difluoride (PVDF) membrane (Millipore, Kilsythe,VIC, Australia) and blocked for 1 hour with 5 % non-fatdry milk in PBS-Tween. Primary antibody for rabbit anti-LPL (1:300, sc-32885 Santa Cruz Biotech, Texas, USA),rabbit anti-HSL (1:150, sc-25843 Santa Cruz Biotech),rabbit anti-EL (1:150, 100030 Cayman chemical, Michigan,USA), or rabbit anti-ATGL (1:300, 2138 Cell SignallingTechnology, Massachusetts, USA were co-incubated withmouse anti-β-Actin (1:20000, A5316, Sigma Aldrich) over-night at 4 °C with agitation. Secondary LI-COR antibodies,goat anti-rabbit 800CW (1:10000, 926–32211, LI-COR)and donkey anti-mouse 680LT (1:15000, 926–68022, LI-COR) were incubated for 1 hour at room temperature andprotein was detected by the Odyssey Infrared Imaging Sys-tem (LI-COR). Lipase protein expression was analyzed bydensitometry correcting for differences in protein loadingby using β-actin levels.

ImmunohistochemistryParaffin-embedded sections (5 μm) were baked, andrehydrated. Antigen retrieval was performed by heat-ing to 125 degrees °C in 100 mM sodium citrate,0.05 % Tween 20 at pH 6.0 for 30 minutes. Endogen-ous peroxidase activity was blocked with hydrogenperoxide 3 % for 10 mins followed by 15 mins withBiocare Background Sniper (MACH2, Biocare Med-ical, Concord CA). Immunolabelling was performedusing polyclonal rabbit antibodies to LPL, anti-HSL,anti-EL antibodies (Biorbyt, Cambridge, UK: LPL(1:1000, orb13546), EL (1:500, orb100394), HSL(1:100, orb40070)) and ATGL antibody (1:100, CellSignalling Technology). Confirmatory immunohisto-chemistry for LPL was performed with a secondpolyclonal rabbit antibody (1:1000, Santa Cruz Bio-tech, sc-32885). After washing, the slides were incubatedwith a biotinylated polyvalent goat secondary antibodyfollowed by DAB incubation for 1 minute. Slides were

counterstained with Harris’ Haematoxylin (HHS 16,Sigma Aldrich) and mounted with coverslips.

Image analysisHSL protein expression was analyzed with a quantita-tive immunohistochemistry method described by Helpset al. [28]. This method uses Ruifrok and Johnston'scolor deconvolution image processing method to digit-ally separate hematoxylin and DAB staining. The im-aging processing and analysis was performed in NIH-ImageJ software using Landini’s ImageJ plugin thenhistogram analysis and a weighting calculation to esti-mate the amount of DAB staining. We took 10 ran-domly selected frames from each of 4 control and 4 PEplacentae that were processed and stained concurrently.Within each frame, the placental villi were demarcatedby hand on the NIH-ImageJ software.

Lipase activityThree milligrams of tissue was homogenized in 300 uLof ice cold assay buffer (150 mM NaCl, 10 mM Tris,2 mM EDTA, pH 7.4) for 4 mins at 30Hz with a stain-less steel bead using the Tissue lyser II (Qiagen). Ho-mogenates were centrifuged for 10 mins at 10000 X g at4 °C. Lipase enzyme activity was measured in superna-tants using a commercial kit (Roar Biomedical, New York,USA) according to the manufacturer’s instructions. LPLactivity in the supernatant was measured in duplicate by afluorescence method as described by the assay manufac-turer. The fluorescence of each sample was normalised tomg of tissue. Measurements were made at baseline, 30and 45 mins of incubation, and the result is expressed aschange from baseline.

LPL promoter Methylation100 ng of genomic DNA was bisulfite treated using theBisulFlash DNA modification kit (Epigentek, USA). Primersfor bisulfite-converted DNA were designed with the onlinetool Methprimer (www.urogene.org//methprimer) covering4 CpG sites in the LPL promoter region from bp −236to −46 prior to the transcription start site of the LPLpromoter. Primer sequences: left primer 5′-TGAGGGAGGATTGTAAGTGATAAATA, right primer 5′- CCCTATCTAAACACCAAACACAAAT. 20 ng bisulfite-treatedDNA was PCR amplified with 1 cycle at 95 °C for10 min, 40 cycles of 95 °C for 30 sec, 55 °C for 40 secand 72 °C for 60 sec, followed by 1 cycle of 7 min at72 °C. The PCR products were then supplied to a se-quencing facility (AGRF, St Lucia QLD Australia) forcapillary sequencing. Sequencing results were analysedwith the online software tool for bisulfite-treated DNABiQ (biq-analyzer.bioinf.mpi-inf.mpg.de/) giving theproportion of methylated and unmethylated residuesat each CpG site.

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 3 of 9

Statistical analysisExperiments were performed in duplicate. Data are pre-sented as mean +/− SEM unless stated otherwise. Differ-ences between groups were examined with two tailedMann–Whitney U tests (Prism version 5.03 software(GraphPad, La Jolla, CA)). Significance was set at < 0.05.Correlation analysis was performed with Spearman’s rhotesting. Sensitivity analyses were performed excludingthe data from the women with SGA infants and no dif-ference was seen from the presented results.

ResultsStudy participantsPlacentae from 16 women with preeclampsia (PE) and20 normotensive (control) women were collected. Thisstudy includes women with late onset rather than earlyonset (pre 32 weeks gestation) PE. The women werematched for maternal BMI and gestational age of deliv-ery. Maternal and pregnancy characteristics are shownin Table 2. There were 4 infants (25 %) born to womenwith PE who were small for gestational age infants(SGA). The mean gestational age of delivery did not dif-fer between groups. The earliest delivery in women withPE was at 35.1 weeks and in the control group was36.6 weeks.

ATGLATGL was localized to syncytiotrophoblasts, endothelialcells and stromal cells including Hofbauer cells anddecidual cells (Fig. 1a, b). The relative expression ofATGL mRNA was reduced in placenta from womenwith PE (PE median 0.15 AU (IQR 0.08–0.32) vs control(1.08 AU (0.30–1.57), P = 0.0003) (Fig. 1c). As the PEgroup included 4 women with SGA infants, we have alsoshown the PE alone and SGA alone results for mRNAexpression in Fig. 1c. Given the smaller numbers, we didnot perform statistical analyses on these different groups.There was no clear difference in the protein expressionbetween placentae from control women or those with PE

for ATGL (PE median 0.84 AU (IQR 0.52–1.20) vs control(0.77 AU (0.55–1.61), P = 0.98) (Fig. 1d, e).

HSLHSL was localized to syncytiotrophoblasts and endothe-lial cells but also to stromal cells including Hofbauercells and decidual cells (Fig. 2a, b). The relative expres-sion of HSL mRNA was reduced in placenta fromwomen with PE (PE median 0.11 AU (IQR 0.06–0.15) vscontrol (0.83 AU (0.30–1.66), P < 0.0001) (Fig. 2c). HSLprotein expression was assessed by quantitative immu-nohistochemistry. There was a reduction in HSL proteinexpression in placentae from women with PE comparedto control (PE median 42.0 %DAB staining (IQR 36.5–48.7), Control 45.0 %DAB staining (39.7–53.6), P <0.0001) (Fig. 2d).

LPL and ELThe extracellular lipases LPL (Fig. 3a, b) and EL (Fig. 4a, b)were localized to syncytiotrophoblasts, endothelial cells andalso to stromal cells including Hofbauer cells and decidualcells. The relative expression of LPL and EL mRNA werereduced in the placenta from women with late onset PE((LPL: PE median 0.13 AU (IQR 0.08–0.33) vs control(0.57 AU (0.26–0.98), P < 0.0001) and EL: PE median0.03 AU (IQR 0.01–0.09) vs control (0.68 AU (0.41–1.67),P < 0.0001)). There was no clear difference in the proteinexpression between placentae from control women or thosewith PE for LPL (PE median 0.63 AU (IQR 0.47–1.16) vscontrol (0.95 AU (0.61–1.33), P = 0.19) (Fig. 3e, f) or EL(PE median 0.68 AU (IQR 0.31–1.81) vs control (0.57 AU(0.30–0.74), P = 0.48) (Fig. 4d, e). DNA methylation of 4CpG sites in the LPL promoter was investigated by bisulfitesequencing. While the levels of methylation varied betweenthe CpG sites, there was no difference in the proportion ofmethylated vs unmethylated CpG sites in the LPL promoterbetween PE and control placentae (Fig. 3g).

Lipase activityLipase activity was measured as fluorescence emission ofhydrolyzed substrate in response to lipase enzyme activ-ity and was tested in homogenized placental tissuesamples. Lipase activity was measured at 30 and 45 mi-nutes at room temperature (Fig. 5). There was nodifference in placental lipase activity between PE andcontrol pregnancies at 30 (PE median 0.39 (IQR 0.36–0.66) vs. control 0.32 (IQR 0.28–0.57), P = 0.12) or 45 mi-nutes (PE median 0.65 (IQR 0.51–1.09) vs. control 0.53(0.47–0.96), P = 0.35).

Relationship with clinical factorsThere was no relationship between the mRNA expres-sion of any lipase and maternal early pregnancy BMI orinfant birth weight for women with or without PE.

Table 2 Clinical characteristics

Control PE P

n 20 16

Maternal age (years) 32.6 (1.0) 31.0 (1.6) 0.40

Maternal BMI in early pregnancy(mean(SD))

26.9 (1.5) 27.6 (1.6) 0.90

Caucasian ethnicity n (%) 19 (95) 16 (100) 1.0

Gestational age of delivery (weeks) 38.7 (0.2) 38.2 (0.4) 0.26

Birth weight (g) 3442 (86.84) 3056 (142.5) 0.06

Birth weight centilea 55.3 (7.2) 38.55 (7.60) 0.15

SGA n(%) b 0 4 (25) 0.03

Infant sex (F/M) 9/11 8/9 1.0aAdjusted birthweight centile [33] badjusted birth centile <10th

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 4 of 9

DiscussionThe current study, examining placentae from women withlate onset PE and BMI matched women with uncomplicatedpregnancies, has demonstrated reduced mRNA expressionand reduced protein expression of HSL in placentae fromwomen with PE. ATGL, LPL and EL showed reducedmRNA expression but unchanged protein expression. Alllipases examined localized to the maternal and fetal sides ofthe placenta as well as the Hofbauer cells and decidual cells.There was no difference in localization between PE andcontrol placentae for any of the four lipases described in thisstudy, but we have demonstrated they are present inplacental cells expected to be metabolically active.The strengths of this study include a well character-

ized and matched cohort, however there are some limi-tations. For example, the HSL protein expression data inthe current study needs to be assessed with some cau-tion. We were unable to obtain western blot results forHSL, suggesting low levels of protein expression of HSL.Human HSL protein concentrations are high in adipo-cytes but much lower in other tissues such as skeletalmuscle [29]. HSL mRNA is clearly present in placenta

and HSL was localised with selective antibody by immu-nohistochemistry. Quantification of protein expressionby image analysis is a validated technique, however, aswith any analysis, it is possible that the reported reduc-tion in HSL is a result of a type 1 error.In late onset PE the fetus is usually well grown. The

presence of a growth restricted fetus suggests that thePE is more severe or that another pathology underliesthe growth restriction [30]. The inclusion of placentaefrom women with growth restricted infants in our studyis therefore a potential confounder. However, we per-formed sensitivity analyses which did not change the dir-ection or significance of any results indicating that thepresence or absence of placentae related to growth re-stricted infants did not influence the results. Our sam-ples were obtained from women with late onset PE only,which could have implications for applying our resultsto general PE. Early PE has been postulated to havequite different placental pathology and function to latePE [31, 30]. The findings of this study should thereforenot be extrapolated to early PE, and the potential differ-ences due to gestational age should be considered.

Fig. 1 ATGL. a Immunohistochemisty for the detection of ATGL in control placenta, open arrow indicates endothelial cell staining, closed arrowindicates trophoblast staining. b immunohistochemistry in PE placenta, bar indicating 50 um. c negative control. d ATGL mRNA expression ofcontrol, all PE, PE alone and PE/SGA cases. e Representative Western blot result. f Relative protein expression of ATGL in control and PE placentae,boxes represent median with interquartile range, whiskers indicating 2.5–97.5 %, *, P < 0.05

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 5 of 9

ATGL and HSL have not previously been examined inplacentae from women with PE. We have recently shownthat ATGL mRNA was increased and HSL mRNA de-creased, with no difference in protein expression, in obesewomen with well controlled gestational diabetes mellituscompared to BMI matched controls [32]. ATGL and HSLare both intracellular lipases. They are involved in themobilization of triacylglycerol from lipid droplets withATGL mainly converting triacylglycerol to diacylglyceroland HSL converting diacylglycerol to monoacylglycerol.The decrease in ATGL and HSL mRNA and HSL proteinseen here could result in reduced lipolysis of placental lipiddroplets and hence reduced lipid transfer to the fetus.In the current study, LPL mRNA expression was re-

duced in placentae from women with PE. However, therewas no difference in protein expression or localisation.LPL mRNA expression has previously been found to beunchanged in PE [21] and unchanged [21] or increasedin growth restricted infants [19, 20]. One reason for thedisparity between our findings and the earlier studies withrespect to growth restricted infants is the gestational ageat delivery. In both the studies showing a reduced LPLmRNA expression in placentae from pregnancies withgrowth restricted infants, the placentae were from preg-nancies delivered at a mean gestational age of 32 weeks at

delivery, whereas delivery was at 39 weeks in the Laivouristudy [21]. The mean gestational age at delivery of the pla-centae in our study is 38 weeks and the lack of changeseen is consistent with the lack of change seen at the latergestation in the Laivouri study. The difference in gesta-tional age rather than the condition itself could underliethe increase in LPL mRNA seen in the studies comparing32 week placenta to term placenta.EL mRNA expression was also reduced in placentae

from women with PE, with no change in protein expres-sion or localisation. EL expression has been previouslyfound to be decreased in growth restricted infant associ-ated placaentae compared with control [19]. It needs tobe noted that this finding was in a cohort comparingterm control and 32 week growth restricted infant re-lated placentae, and that they reported an increase inplacental EL expression from the first to third trimester.Once again, in our study, with careful matching of gesta-tional age, EL protein expression was unchanged in PE.The current study found no difference in overall lipase

activity measured in placental biopsies. This is consistentwith the results of a previous study in preterm and termpregnancies complicated by growth restricted infantsthat reported no change in overall placental triglyceridehydrolase activity but reduced LPL activity in isolated

Fig. 2 HSL. a Immunohistochemisty for the detection of HSL in control placenta, open arrow indicates endothelial cell staining, closed arrowindicates trophoblast staining. b immunohistochemistry in PE placenta. c negative control. d mRNA expression, showing control compared withall PE cases with the result of statistical analysis, as well as the result for PE alone or PE/SGA cases. e Relative protein expression as measured bysemi-quantitative immunohistochemistry. *, P < 0.05

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 6 of 9

Fig. 3 LPL. a Immunohistochemisty for the detection of LPL in control placenta, open arrow indicates endothelial cell staining, closed arrowindicates trophoblast staining. b immunohistochemistry in PE placenta. c negative control. d mRNA expression, showing control compared withall PE cases with the result of statistical analysis, as well as the result for PE alone or PE/SGA cases. e Representative Western blot result. f Relativeprotein expression, boxes represent median with interquartile range, whiskers indicating 2.5–97.5 %, *, P < 0.05. g LPL methylation analysis

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 7 of 9

placental microvillous membrane [18]. In contrast, anolder study showed greater LPL but lower intracellularlipase activity in placentae from PE and IUGR pregnancy[22]. The assay used in the current study is conducted atpH 7 and measures overall triglyceride hydrolase activity.The similar lipase activity levels we report are in keepingwith the lack of change in protein levels for 3 of 4 lipaseswe examined, suggesting that in term PE, placental li-pases are unaltered.

ConclusionThe current study demonstrated a decrease in mRNAexpression in all four lipases. A small decrease in HSLprotein was seen, but no changes in protein expressionfor ATGL, LPL or EL were demonstrated. There was nodifference in lipase activity in placentae from pregnan-cies complicated by late onset PE compared to control.This suggests that this aspect of placental lipid process-ing is not altered in late onset PE and does not underliethe differences seen in infant growth.

Fig. 4 EL. a Immunohistochemisty for the detection of EL in control placenta, b immunohistochemistry in PE placenta, open arrow indicatesendothelial cell staining, closed arrow indicates trophoblast staining. . c negative control. d mRNA expression, showing control compared withall PE cases with the result of statistical analysis, as well as the result for PE alone or PE/SGA cases. e Representative Western blot result, boxesrepresent median with interquartile range, whiskers indicating 2.5–97.5 %, *, P < 0.05. f Relative protein expression

Fig. 5 Lipase activity. Change in placental lipase activity taken at 30and 45 mins, intissue from control pregnancies in white, and PE ingrey. n = 8 control and n = 14 PE. Corrected to mg of tissue

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 8 of 9

Competing interestsThe authors declare that they have no competing interests.

Authors’ contributionsHLB conceived of project, obtained funding, performed analyses andinterpreted data, wrote and edited manuscript, MK KSR performed analysesand edited manuscript, TR KD obtained funding, edited the manuscript,HDM conceived of project, obtained funding, interpreted data, editedmanuscript, LKC conceived of project, obtained funding, interpreted data,edited manuscript, MDN conceived of project, obtained funding, performedanalyses and interpreted data, edited manuscript. All authors approved thefinal version of the manuscript.

FundingProject funding for this analysis was obtained from Pfizer Australia with aCardiovascular Lipid Research grant, a National Health and Medical ResearchCouncil project grant (APP 569693), and a Royal Brisbane and Women’sHospital foundation grant. HLB is supported by a National Health andMedical Research Council medical and dental PhD scholarship. MDN issupported by a Patricia Dukes Fellowship from a Royal Brisbane andWomen’s Hospital foundation grant.

Author details1UQ Centre for Clinical Research, The University of Queensland, Herston,QLD, Australia. 2Obstetric Medicine, Royal Brisbane and Women’s Hospital,Herston, QLD, Australia. 3School of Medicine, The University of Queensland,Herston, QLD, Australia. 4Mater Research Institute, The University ofQueensland Brisbane, Brisbane, QLD, Australia. 5School of BiomedicalSciences, The University of Queensland, St Lucia, QLD, Australia.

Received: 23 June 2015 Accepted: 20 August 2015

References1. van Pampus MG, Aarnoudse JG. Long-term outcomes after preeclampsia.

Clin Obstet Gynecol. 2005;48(2):489–94.2. Goynumer G, Yucel N, Adali E, Tan T, Baskent E, Karadag C. Vascular risk in

women with a history of severe preeclampsia. JCU. 2013;41(3):145–50.doi:10.1002/jcu.21962.

3. Mayer C, Joseph KS. Fetal growth: a review of terms, concepts and issuesrelevant to obstetrics. Ultra Obstet Gynecol. 2013;41(2):136–45.doi:10.1002/uog.11204.

4. Chan PY, Morris JM, Leslie GI, Kelly PJ, Gallery ED. The long-term effects ofprematurity and intrauterine growth restriction on cardiovascular, renal, andmetabolic function. Int J Pediatr. 2010;2010:280402. doi:10.1155/2010/280402.

5. Fraser A, Nelson SM, Macdonald-Wallis C, Sattar N, Lawlor DA. Hypertensivedisorders of pregnancy and cardiometabolic health in adolescent offspring.Hypertension. 2013;62(3):614–20. doi:10.1161/hypertensionaha.113.01513.

6. Ziaei S, Bonab KM, Kazemnejad A. Serum lipid levels at 28–32 weeksgestation and hypertensive disorders. Hypertens Pregnancy. 2006;25(1):3–10.doi:10.1080/10641950500543756.

7. Sep S, Rijvers C, Smits L, van Bilsen M, Bekers O, Peeters L. Early-pregnancychanges in maternal lipid profile in women with recurrent preeclampsiaand previously preeclamptic women with normal next pregnancy. ReprodSci. 2011;18(10):998–1004. doi:10.1177/1933719111401665.

8. Ray JG, Diamond P, Singh G, Bell CM. Brief overview of maternaltriglycerides as a risk factor for pre-eclampsia. BJOG. 2006;113(4):379–86.doi:10.1111/j.1471-0528.2006.00889.x.

9. Sattar N, Bendomir A, Berry C, Shepherd J, Greer IA, Packard CJ.Lipoprotein subfraction concentrations in preeclampsia: pathogenicparallels to atherosclerosis. Obstet Gynecol. 1997;89(3):403–8.doi:10.1016/s0029-7844(96)00514-5.

10. Ogura K, Miyatake T, Fukui O, Nakamura T, Kameda T, Yoshino G. Low-densitylipoprotein particle diameter in normal pregnancy and preeclampsia. JAtheroscler Thromb. 2002;9(1):42–7.

11. Alvino G, Cozzi V, Radaelli T, Ortega H, Herrera E, Cetin I. Maternal and fetalfatty acid profile in normal and intrauterine growth restriction pregnancieswith and without preeclampsia. Pediatr Res. 2008;64(6):615.

12. Adiga U, D'Souza V, Kamath A, Mangalore N. Antioxidant activity and lipidperoxidation in preeclampsia. J Chin Med Assoc. 2007;70(10):435–8.doi:10.1016/S1726-4901(08)70034-0.

13. Bayhan G, Kocyigit Y, Atamer A, Atamer Y, Akkus Z. Potential atherogenicroles of lipids, lipoprotein(a) and lipid peroxidation in preeclampsia. GynecolEndocrinol. 2005;21(1):1–6. doi:10.1080/09513590500097382.

14. Magnussen EB, Vatten LJ, Smith GD, Romundstad PR. Hypertensive disordersin pregnancy and subsequently measured cardiovascular risk factors. ObstetGynecol. 2009;114(5):961–70. doi:10.1097/AOG.0b013e3181bb0dfc.

15. He S, Silveira A, Hamsten A, Blomback M, Bremme K. Haemostatic, endothelialand lipoprotein parameters and blood pressure levels in women with a historyof preeclampsia. Thromb Haemost. 1999;81(4):538–42.

16. Desoye G, Gauster M, Wadsack C. Placental transport in pregnancypathologies. Am J Clin Nutr. 2011. doi:10.3945/ajcn.110.000851.

17. Gil-Sanchez A, Koletzko B, Larque E. Current understanding of placental fattyacid transport. Curr Opin Clin Nutr Metab Care. 2012;15(3):265–72.doi:10.1097/MCO.0b013e3283523b6e.

18. Magnusson AL, Waterman IJ, Wennergren M, Jansson T, Powell TL. Triglyceridehydrolase activities and expression of fatty acid binding proteins in the humanplacenta in pregnancies complicated by intrauterine growth restriction anddiabetes. J Clin Endocrinol Metab. 2004;89(9):4607–14. doi:10.1210/jc.2003-032234.

19. Gauster M, Hiden U, Blaschitz A, Frank S, Lang U, Alvino G, et al.Dysregulation of placental endothelial lipase and lipoprotein lipase inintrauterine growth-restricted pregnancies. J Clin Endocrinol Metab.2007;92(6):2256–63. doi:10.1210/jc.2006-2403.

20. Tabano S, Alvino G, Antonazzo P, Grati FR, Miozzo M, Cetin I. Placental LPL geneexpression is increased in severe intrauterine growth-restricted pregnancies.Pediatr Res. 2006;59(2):250–3. doi:10.1203/01.pdr.0000199441.62045.a1.

21. Laivuori H, Gallaher MJ, Collura L, Crombleholme WR, Markovic N,Rajakumar A, et al. Relationships between maternal plasma leptin,placental leptin mRNA and protein in normal pregnancy, pre-eclampsiaand intrauterine growth restriction without pre-eclampsia. Mol HumReprod. 2006;12(9):551–6. doi:10.1093/molehr/gal064.

22. Biale Y. Lipolytic activity in the placentas of chronically deprived fetuses.Acta Obstet Gynecol Scand. 1985;64(2):111–4.

23. Lindegaard ML, Damm P, Mathiesen ER, Nielsen LB. Placental triglycerideaccumulation in maternal type 1 diabetes is associated with increased lipasegene expression. J Lipid Res. 2006;47(11):2581–8. doi:10.1194/jlr.M600236-JLR200.

24. Lager S, Magnusson-Olsson A, Powell T, Jansson T. Placental lipoprotein lipaseactivity is decreased by elevated free fatty acids. Placenta. 2006;27(9–10):A35-A.

25. Magnusson-Olsson AL, Lager S, Jacobsson B, Jansson T, Powell TL. Effectof maternal triglycerides and free fatty acids on placental LPL incultured primary trophoblast cells and in a case of maternal LPLdeficiency. Am J Physiol Endocrinol Metab. 2007;293(1):E24–30.doi:10.1152/ajpendo.00571.2006.

26. Gauster M, Hiden U, van Poppel M, Frank S, Wadsack C, Hauguel-deMouzon S, et al. Dysregulation of placental endothelial lipase in obesewomen with gestational diabetes mellitus. Diabetes. 2011;60(10):2457–64.doi:10.2337/db10-1434.

27. Lowe SA, Brown MA, Dekker GA, Gatt S, McLintock CK, McMahon LP, etal. Guidelines for the management of hypertensive disorders ofpregnancy 2008. Aust N Z J Obstet Gynaecol. 2009;49(3):242–6.doi:10.1111/j.1479-828X.2009.01003.x.

28. Helps SC, Thornton E, Kleinig TJ, Manavis J, Vink R. Automatic nonsubjectiveestimation of antigen content visualized by immunohistochemistry using colordeconvolution. AIMM. 2012;20(1):82–90. doi:10.1097/PAI.0b013e31821fc8cd.

29. Alsted TJ, Nybo L, Schweiger M, Fledelius C, Jacobsen P, Zimmermann R, etal. Adipose triglyceride lipase in human skeletal muscle is upregulated byexercise training. Am J Physiol Endocrinol Metab. 2009;296(3):E445–53.doi:10.1152/ajpendo.90912.2008.

30. Redman CW, Sargent IL, Staff AC. IFPA Senior Award Lecture: Making senseof pre-eclampsia - Two placental causes of preeclampsia? Placenta.2014;35(Suppl):S20–5. doi:10.1016/j.placenta.2013.12.008.

31. Raymond D, Peterson E. A critical review of early-onset and late-onsetpreeclampsia. Obstet Gynecol Surv. 2011;66(8):497–506.doi:10.1097/OGX.0b013e3182331028.

32. Barrett HL, Kubala MH, Scholz Romero K, Denny KJ, Woodruff TM, McIntyre HD,et al. Placental lipases in pregnancies complicated by gestational diabetesmellitus (GDM). PLoS One. 2014;9(8), e104826. doi:10.1371/journal.pone.0104826.

33. GROW Customised Weight Centile Calculator, version 6.5, Australianpopulation [database on the Internet]. Gestation Network www.gestation.net.2013. Accessed:

Barrett et al. Reproductive Biology and Endocrinology (2015) 13:100 Page 9 of 9